Review of Randomized Trials

This is Part 2 of a four-part article. Part 1 provides a general introduction for the whole article which is available as a single PDF file on ResearchGate. The article was presented at the University of Ottawa on November 21, 2015, and is available in video as Parts One and Two.
PART 2: Review of the randomized trials for treatments and screening
Here, I critically review the randomized trials for treatments and screening, especially for breast cancer. It has been advanced that screening does more harm than good, and that treatment protocols have little effect on net population mortality from cancer. There is no robust demonstration that the treatment protocols for the common cancers do more good than harm to individual patients.
This very month I received a personal letter to my home from a vice-president at “Cancer Care Ontario” inviting me to “get checked for colon cancer”. The Canadian province of Ontario has state-funded screening programs for breast, cervical, and colorectal cancers. The program logo says: “Screen for Life. Cancer screening sees what you can’t.” (italics in the original). The colour-flyer supplement directs me to “Get informed about this important health initiative”, but it does not contain any information or provide a link to information that would allow me to judge “this important health initiative”. What is the science that justifies these government programs, and how reliable is that science?
There is a relation between screening and treatments (surgery, chemo, radiation, and various adjuvant therapies) because screening is only beneficial if treatments are sufficiently effective. To be critical of screening programs is somewhat to pooh-pooh the treatments. As a result, the insider policy debates have been intense (see below).
I have chosen to concentrate on breast cancer, which is classified as the leading life-threatening disease in women, when tumours are also present in other organs. As I did the reading, I was continually amazed at the plethora of false logic and researchers’ trust in dubious methods and tenuous statistics. Here are some examples.
For a start, I was not able to find even a reference to a study that established that surgery was of any value in treating breast cancer. Quite simply, practitioners are so convinced that to remove a breast tumour must be beneficial that a rigorous scientific evaluation of this point is never entertained; so much for “evidence-based medicine”.
For example, in 1961, Goldenberg et al. put it this way1:

… the first recorded description [of breast cancer] was in an Egyptian papyrus written about 1500 B.C. Operative therapy was not recognized then; rather, it was left to the internist to apply caustic ointments locally. A great deal of progress was made during succeeding centuries so that by 1600, standard therapy was elevation of the diseased breast by large pincers, amputation with a knife and application of a poker heated to red intensity to the massive wound for hemostasis. In 1746, Angelo Nannoni published his Surgical Treatise on Diseases of the Breast in which he advocated, for the first time, removal of the breast, underlying fascia, pectoralis major muscle and “hardened axillary glands” as soon as the surgeon recognized a possible malignant process. The basis was thereby established for operative treatment of breast cancer. … but it was not until the 1890’s that Halsted and Willy Meyer almost simultaneously described radical mastectomy as the operation of choice. Little has been added to the operative procedure since that time.” [p. 397] “… Radical mastectomy has been standard therapy, not so much because patients are always cured by this operation, but rather because surgeons do not know what else to do. [p. 402]

These authors went on to describe their study of 1,458 breast cancer patients. They concluded that surgery improved survival (i.e., avoiding death in the years following the first diagnostic). However, this conclusion is unreliable since, in their own words: “…there are always some patients who receive no treatment at all. This latter group is, fortunately, small and consists mainly of those who refuse recommended therapy.” [p. 402]. Thus, this study, which became a cited source claiming that surgery is beneficial, used a small unspecified number of untreated patients, which were not selected by randomization. No double-blind, no comparative study, not even randomization, on a “small” sample.
Fisher et al. cited Goldenberg et al. in 1969 in exclaiming2:

That earlier diagnosis, i.e., the discovery of smaller tumors, will result in improved survival has been categorically accepted, and evidence from the literature tends to support this consideration. Indeed, the concept upon which all neoplastic surgery is predicated is that time—considered synonymous with tumor size—is the important factor between localized and disseminated disease, and that operations performed at the earliest possible time give the best chance of cure.  [All without evidence, of course.]

Fisher et al. went on to explain:

Surgery for cancer has been and is predicated on the concept that, a. a growing tumor remains localized for a period of time, b. at some instant during its growth, depending upon the tumor type, tumor cell dissemination to regional nodes begins to take place, c. after a further time interval associated with continued increase in tumor size, systemic dissemination ensues, and d. adequate surgery performed prior to the latter event ensures cure.

In other words, the dominant paradigm of the disease (i.e., metastasis, see Part-III below) is so psychologically compelling that researchers, let alone clinicians, don’t need evidence.
Thus, surgery alone, as a treatment, is never evaluated and its supposed benefits have never been demonstrated. Only so-called adjuvant (“preventative”) therapies are considered, in combination with tumour removal or ablation. Again in 1969, the prominent researcher Fisher wrote3:

The concept of administering a chemo-therapeutic agent systemically in conjunction with radical mastectomy to decrease the recurrence rate and enhance survival of patients with breast cancer is an appealing one. [Really?]
As a result of publicity accorded these studies at their onset – before their worth or danger was remotely established – adjuvant chemotherapy was widely employed by many in the treatment of breast cancer. Only recently have definitive results of this project become known. [Emphasis added.]
At the end of 5 years, there was no significant difference in recurrence rate between patients receiving TSPA or placebo … Fifty-two percent of patients who received 5-FU suffered local complications, and significantly more patients receiving that drug demonstrated systemic complications. Moreover, 42 percent of patients administered 5-FU became leukopenic and almost half of those had counts < 2,500. Of the 16 deaths occurring within 60 days following surgery in the 1,725 patients of this second study, 8 were recipients of 5-FU and all demonstrated toxic manifestations of the drug.

It is a wonder that lawsuits did not cause the science to become more rigorous, and that highly toxic substances would continue to be prescribed to this day, without any robust demonstrations of net benefits.
A massive government survey published in 1981 made the following common sense conclusion about cancer treatments4:

Consequently, as we are not convinced that changes in treatment have materially affected the outcome of most of the major types of potentially fatal cancer, it seems to us wiser for most types of cancer to estimate the real trends in disease onset rates chiefly from the recorded trends in mortality since 1950 among people under the age of 65.

In other words, forget all the claims of treatment benefits and look at hard numbers of deaths from cancer, if you want an objective measure.
Following this criterion, I have looked for robust demonstrations of benefits, based on actual comparative studies without bias (not simple “randomization” without proper controls; with double-blind treatment dispensation and reporting; using arms-length evaluations of diagnoses and outcomes)—there are none.
Next, I have tried to identify the most cited papers, and those cited in the most prestigious reviews. I consistently found that the data interpretations are dubious, that the conclusions of statistical significance are overly optimistic, and that the study methods and designs are susceptible to large bias. (References5,6,7,8,9,10,11,12, and13 are a sample.)
In one case, I followed a large randomization-trial study about alleged benefits of screening, throughout its published history.14,15,16,17 Such trials have a large potential for bias, and require difficult determinations of the cause of death, where the numbers of deaths from all other causes are approximately ten times greater than the numbers of deaths from cancer. This study reported a staggering 30% difference in mortality at 7 years following “randomization” (i.e., 7 years following invitation to be screened). I found inconsistencies in the articles and asked the lead author for explanations. The answers from him and a co-author did not alleviate my concerns, and suggested that, at best, important methodological features had not been specified in the published articles.
As the latter randomization-trial study was coming to completion, Jørgensen and Gøtzsche found that mammography screening programmes have an overdiagnosis rate of 52%.18 “Overdiagnosis” is “the detection of cancers that will not cause death or symptoms”. This means that 46% to 58% (95% confidence interval) of the patients unnecessarily suffer all the inconveniences, stresses, and health harms directly resulting from diagnoses and treatments.
Recently, the renowned expert Peter Gøtzsche has firmly concluded “mammography screening is harmful and should be abandoned”, and a nationally appointed body in Switzerland has recommended that the country stop its breast cancer screening program because it is harmful.19 In the end, all those “randomized” trials supposedly proving that there are benefits to screening for breast cancer are probably a lot of hogwash. This is not surprizing in the light of the studies reviewed in Part-I, above.
This does not mean that those patients that did have symptoms and died benefitted from the treatments. It only means that half of the screened patients were aggressed by the medical system, at great cost, for no valid reason. The question of whether there is ever any benefit from the accepted treatments, which does not arise purely from the psychological context of the treatments, is an open question that cannot be answered in the affirmative.
An objective and robust measure of treatment benefit lies in the death rates from breast cancer. Even given the system bias in wanting these reported rates to decrease as treatments are “improved” (and as Big Pharma profits soar), the death rates cannot be related to changes in treatments.20
In conclusion, breast cancer oncologists prescribe poison, radiation, and invasive surgery with little regard for the harm caused by their diagnoses and “treatments”. The history of the “treatments” is gruesome and the professional culture is antithetical to evidence-based science, despite the lip service.
Furthermore, there is a large inertia against self-examination and self-criticism, as can be seen from any lofty professional association statement21:

Quality of life needs to be evaluated in selected randomized clinical trials to examine the impact of the major acute and long-term side effects of adjuvant treatments, particularly premature menopause, weight gain, mild memory loss, and fatigue. Methods to support shared decision-making between patients and their physicians have been successful in trials; they need to be tailored for diverse populations and should be tested for broader dissemination.

This is where we have ended up: “We need clinical trials to evaluate the “side” effects of toxic substances that have not been shown to be of net benefit…” (my words).
Parts 3 and 4
In Part 3 of the whole article I will critically review the mutation-centric metastasis dominant paradigm of cancer, and various efforts to somewhat or definitively challenge the dominant paradigm, with an eye to answering the question “What is cancer?”
In Part 4, I will propose a conceptual model of cancer (“age-dependent and tissue-specific stress-induced breakdown of tissue-shape homeostasis”), which incorporates the leading criticisms of the dominant paradigm.

1. I.S. Goldenberg et al. “Female Breast Cancer: A Re-evaluation” Annals of Surgery, September 1961, vol. 154, no. 3, pages 397-404.
2. B. Fisher et al. Cancer of the breast: Size of neoplasm and prognosis. Cancer, November 1969, vol. 24, no. 5, pages 1071-1080.
3. B. Fisher. Systematic chemotherapy as an adjuvant to surgery in the treatment of breast cancer. Cancer, December 1969, vol. 24, no. 6, pages 1286-1289.
4. R. Doll and R. Peto. The Causes of cancer: Quantitative estimates of avoidable risks of cancer in the United States today. Journal of the National Cancer Institute, June 1981, vol. 66, no. 6, pages 1192-1308.
5. D.M. Parkin. Cancer of the Breast, Endometrium and Ovary: Geographic Correlations. European Journal of Cancer and Clinical Oncology, 1989, vol. 25, no. 12, pages 1917-1925
6. C.L. Carter. Relation of Tumor Size, Lymph Node Status, and Survival in 24,740 breast Cancer Cases. Cancer, 1 January 1989, vol. 63, no. 1, pages 181-187
7. Early Breast Cancer Trialist’s Collaborative Group. Systemic treatment of early breast cancer by hormonal, cytotoxic, or immune therapy: 133 randomized trials involving 31,000 recurrences and 24,000 deaths among 75,000 women. Lancet, 4 January 1992, vol. 339, no. 8784, pages 1-15
8. R. Peto. Five Years of Tamoxifen—or More? (Editorial). Journal of the National Cancer Institute, 18 December 1996, vol. 88, no. 24, pages 1791-1793
9. R.G. Blanks et al. Effect of NHS breast screening programme on mortality from breast cancer in England and Wales, 1990-8: comparison of observed with predicted mortality. British Medical Journal (BMJ), 16 September 2000, vol. 321, no. 7262, pages 665-669
10. J. McCann et al. Predicted long-term mortality reduction associated with the second round of breast screening in East Anglia. British Journal of Cancer, 2001, vol. 84, no. 3, pages 423-428
11. Early Breast Cancer Trialist’s Collaborative Group. Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomized trials. Lancet, 14 may 2005, vol. 365, pages 1687-1717.
12. D.A. Berry et al. Effect of Screening and Adjuvant Therapy on Mortality from Breast Cancer. New England Journal of Medicine (NEJM), 27 October 2005, vol. 353, no. 17, pages 1784-1792
13. M. Dowsett et al. Meta-Analysis of Breast Cancer Outcomes in Adjuvant Trials of Aromatase Inhibitors Versus Tamoxifen. Journal of Clinical Oncology, 20 January 2010, vol. 28, no. 3, pages 509-518
14. L. Tabar et al. Reduction in mortality from breast cancer after mass screening with mammography: Randomized trial from the breast cancer screening working group of the Swedish National Board of health and Welfare. Lancet, 13 April 1985, pages 829-832
15. L. Tabar et al. The Swedish two county trial of the mammographic screening for breast cancer: recent results and calculation of benefit. Journal of Epidemiology and Community Health, June 1989, vol. 43, no. 2, pages 107-114
16. L. Tabar et al. The Natural History of Breast Carcinoma: What Have We Learned from Screening? Cancer, 1 August 1999, vol. 86, no. 3, pages 449-462
17. L. Tabar et al. Swedish Two-County Trial: Impact of Mammographic Screening on Breast Cancer Mortality during 3 Decades. Radiology, September 2011, vol. 260, pages 658-663
18. K.J. Jørgensen and P.C. Gøtzsche. Overdiagnosis in publicly organized mammography screening programmes: systematic review of incidence trends. British Medical Journal (BMJ), 2009, vol. 339:b2587 doi:10.1136/bmj.b2587 – 8 pages
19. P.C. Gøtzsche. Mammography screening is harmful and should be abandoned. Journal of the Royal Society of Medicine, 2015, vol. 108, no. 9, pages 341-345. doi: 10.1177/0141076815602452
20. Ibid., Figure 2; and see J.L. Botha et al. Breast cancer incidence and mortality trends in 16 European countries. European Journal of Cancer, 2003, vol. 39, pages 1718-1729
21. National Institutes of Health Consensus Development Panel. National Institutes of Health Consensus Development Conference Statement: Adjuvant Therapy for Breast Cancer, November 1-3, 2000. Journal of the National Cancer Institute, 4 July 2001, vol. 93, no. 13, pages 979-989